Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     definition
    -0.65
    gans
    -0.65
     silenced
    -0.64
     whole
    -0.62
     parole
    -0.61
    âĦ¢:
    -0.61
     capt
    -0.60
     hunger
    -0.59
     censored
    -0.59
     penetrating
    -0.59
    POSITIVE LOGITS
    estate
    0.71
    imb
    0.70
    agnar
    0.69
     Proceed
    0.68
    orm
    0.68
     Mehran
    0.68
    ords
    0.65
    ormonal
    0.65
    imer
    0.63
    trop
    0.63
    Act Density 0.000%

    No Known Activations

    This feature has no known activations.